Data tags are becoming a standard convention for data from feeds. They appear on web sites, etc., and much of the tagging of data from such sites is actually being performed by users of the sites who end up “donating” the information while signing up for the tagging services. However, different sets from different sources (i.e., different web sites) may have different agreements or conventions about tag naming. When a person receives data feeds from a variety of sources, the different data sets may have a disparate variety of tags that do not identify the data types consistently.
What is clearly needed is a system and method that can help automatically build a knowledge base for enriching feeds coming from different sources and that have tags of different conventions, by deducting which tags go into various categories of knowledge.
As it is possible to associate tags, it may also be useful and interesting to know how the items to which the tags are attached are related to one another.
What is clearly further needed is a system and method for deducing the relationship between objects, based on the relationships between their tags.